Despite recent evidence demonstrating treatment-induced neuroplasticity in chronic aphasia, questions remain regarding the influence of treatment variables on language recovery, the neural mechanisms supporting such recovery, and evidence of measurable, correlated improvements in quality of life (QoL). The long-range goal of the proposed study is to better understand how treatment variables affect recovery in chronic aphasia, including the brain/behavior mechanisms supporting improved naming, and how such improvements impact QoL. The current objective is: first, to determine whether intensive treatment for remediation of word-finding deficits is effective for persons with moderate-to-severe expressive aphasia, regardless of speech practice; second, to identify signature neural patterns of treatment-induced improvements in naming; and third, to characterize the relationship between improved language ability and QoL within affected families. The current study proposes a longitudinal case series approach to aphasia rehabilitation, focusing on the effect of one variable, implicit vs. explicit treatment, on linguistic, neurophysiological, and psychosocial changes in chronic moderate-to-severe patients. The three specific aims are: 1) To determine whether gains in naming following constraint-induced language therapy (CILT) exceed those observed following unconstrained language therapy (ULT), over and above type and severity of aphasia, and whether the gains can be maintained over time. The central hypothesis is that participants, even those with persistent severe nonfluent aphasia and co-morbid apraxia of speech, will benefit from both intensive treatment programs, with significant gains in those randomly assigned to CILT. Words that are trained and practiced (i.e., overlearned) following CILT and ULT will show better maintenance in 12-month follow-up testing; 2) To characterize patterns of fMRI activation that correlate with improved picture naming. The working hypothesize is that improved naming correlates with modulation of spared networks that vary according to site/extent of lesion; and 3) To quantify the degree to which improvements following short-term intensive language therapy correlate with changes in QoL measures as perceived by both participants with aphasia and their significant others. Preliminary data suggest that improved naming can jumpstart language systems caught in learned non-use and stimulate behavioral changes in factors known to correlate with QoL, e.g., confidence, social relationships, and independence. This research is innovative, first because of the three-pronged (behavioral, neuroimaging, QoL) longitudinal case series approach and its potential to demonstrate meaningful long-term changes resulting from short-term, intensive treatment plus a maintenance home therapy program, and second, because it focuses, in part, on a more severely affected population than is customarily studied. This contribution is significant because advancing our knowledge of the true anatomical, functional, and behavioral limits and potential for post-stroke language recovery could influence perceptions that affect delivery of services to chronically aphasic persons.